RESUMO
Fluctuating sensorineural hearing lossï¼FSNHLï¼ is a special type of sensorineural hearing loss, which can be manifested in many clinical diseases. In this paper, some clinical diseases associated with FSNHL are summarized, such as Meniere's disease, large vestibular aqueduct syndrome, acute low frequency sensorineural hearing loss, delayed endolymphatic hydrops, autoimmune inner ear disease and syndromes leading to FSNHL. The pathogenesis, diagnosis and treatment of FSNHL were summarized in order to improve clinicians' understanding of FSNHL, reduce the probability of misdiagnosis of related diseases and improve the prognosis of patients.
Assuntos
Hidropisia Endolinfática , Perda Auditiva Neurossensorial , Doença de Meniere , Humanos , Hidropisia Endolinfática/complicações , Doença de Meniere/complicações , Perda Auditiva Neurossensorial/etiologia , Síndrome , PrognósticoRESUMO
In this paper, we do research on cross-corpus speech emotion recognition (SER), in which the training and testing speech signals come from different speech corpus. The mismatched feature distribution between the training and testing sets makes many classical algorithms unable to achieve better results. To deal with this issue, a transfer learning and multi-loss dynamic adjustment (TLMLDA) algorithm is initiatively proposed in this paper. The proposed algorithm first builds a novel deep network model based on a deep auto-encoder and fully connected layers to improve the representation ability of features. Subsequently, global domain and subdomain adaptive algorithms are jointly adopted to implement features transfer. Finally, dynamic weighting factors are constructed to adjust the contribution of different loss functions to prevent optimization offset of model training, which effectively improve the generalization ability of the whole system. The results of simulation experiments on Berlin, eNTERFACE, and CASIA speech corpora show that the proposed algorithm can achieve excellent recognition results, and it is competitive with most of the state-of-the-art algorithms.
Assuntos
Emoções , Fala , Algoritmos , Aprendizagem , Aprendizado de MáquinaRESUMO
Objective:To explore the application value of video head impulse testï¼vHITï¼, caloric testï¼CTï¼ and the dizziness handicap inventoryï¼DHIï¼ in the diagnosis of acoustic neuromaï¼ANï¼, to analyze the correlation between vHIT and CT, and to determine the correlationsof tumor size, vHIT, CT and DHI score. Methods:The clinical data of 24 patients with AN who underwent surgery in our department from January 2019 to January 2022 were analyzed retrospectively, including craniocerebral MRI, vHIT, caloric test and DHI score. All the data were statistically analyzed by GraphPadPrism9.0. Results:There was a significant negative correlation between the UW value of CT and the vestibular eye reflex gain of vHITï¼P<0.01, r=-0.62ï¼. The tumor size was significantly correlated with the increase of UW value of CTï¼P<0.01, r=0.69ï¼, and with the decrease of vestibulo-ocular reflex gain of vHITï¼P<0.01, r=-0.53ï¼. The average Dizziness Handicap Inventory score was 8.9±16.2, which was not correlated with tumor sizeï¼P>0.05ï¼. Conclusion:Both vHIT and CT can effectively evaluate the vestibular function of patients with ANï¼and they are complementaryï¼, and they are related to the size of the tumor and have certain value in the diagnosis of acoustic neuroma.